Virtual petroleum system

Abstract

A method of stochastically modeling a plurality of litho-facies within a formation includes defining a fades classification for each of a top and a base of the formation, dividing the formation into a plurality of layers, and interpolating classifications for each of the plurality of layers, based on the defined facies classification for the top and the base, wherein the interpolating includes a random variation component.

1 . A method of stochastically modeling a plurality of litho-facies within a formation comprising:
 defining a facies classification for each of a top and a base of the formation;   dividing the formation into a plurality of layers; and   interpolating classifications for each of the plurality of layers, based on the defined facies classification for the top and the base, wherein the interpolating includes a random variation component.   
   
   
       2 . A method as in  claim 1 , wherein the random variation comprises a normal distribution. 
   
   
       3 . A method as in  claim 1 , wherein the interpolating includes a component having a gradual variation, from the classification of the top to the classification of the bottom and the random variation component. 
   
   
       4 . A method as in  claim 3 , wherein the gradual variation comprises a weighted average of a composition of the top and the base, wherein the weighting is dependent on a distance of the layer from the top and base. 
   
   
       5 . A method as in  claim 1 , wherein for each layer, a lateral distribution of classifications along the layer is varied in accordance with a normal distribution. 
   
   
       6 . A method as in  claim 1 , wherein for each layer, a lateral distribution of classifications along the layer is varied in accordance with an iterated normal distribution. 
   
   
       7 . A method as in  claim 1 , wherein for each layer, a lateral distribution of classifications along the layer is varied in accordance with an iterated random distribution. 
   
   
       8 . A method as in  claim 7 , wherein the random distribution is geophysically constrained. 
   
   
       9 . A method as in  claim 8 , wherein the geophysical constraint comprises user-applied constraints. 
   
   
       10 . A method as in  claim 8 , wherein the geophysical constraint comprises information derived from geophysical measurements of the formation. 
   
   
       11 . A method as in  claim 1 , wherein the classifications include distributions of material types and wherein between layers, a sum of a fraction of each type is kept the same as a user-defined value. 
   
   
       12 . A system for stochastically modeling a plurality of litho-facies within a formation comprising:
 a data storage system, configured and arranged to store data relating to a plurality of characteristics of a geological region; and   a modeling module, configured and arranged to:
 process the stored data and to produce modeled attributes of at least a portion of the geological region; 
 define a facies classification for each of a top and a base of the formation; 
 divide the formation into a plurality of layers; and 
 interpolate classifications for each of the plurality of layers, based on the defined facies classification for the top and the base, wherein the interpolating includes a random variation component. 
   
   
   
       13 . A system as in  claim 12 , wherein the modeling module divides the formation and interpolates classifications based, at least in part, on user input. 
   
   
       14 . A system as in  claim 12 , wherein the modeling module interpolates classifications based at least in part on a component having a gradual variation from the classification of the top to the classification of the bottom and the random variation component and wherein the gradual variation comprises a weighted average of a composition of the top and the base, wherein the weighting is dependent on a distance of the layer from the top and base. 
   
   
       15 . A system as in  claim 12 , wherein the modeling module interpolates classifications such that for each layer, a lateral distribution of classifications along the layer is varied in accordance with an iterated normal distribution.